Dramatic insider warnings over AI fall flat with some in Silicon Valley

What the report says

At a Goldman Sachs conference in San Francisco this week, some of Silicon Valley’s biggest AI figures confronted a wave of public warnings from current and former industry insiders about the risks of advanced artificial intelligence. The BBC reported that the latest alarm came after Anthropic researcher Jacob Coxon resigned and said people building AI believed it could threaten humanity. Other Anthropic staff have also publicly raised similar concerns, including claims that AI could someday cause catastrophic harm.

According to the BBC, those warnings were met with growing scepticism among executives and investors. Some critics suggested the language may be exaggerated or even designed to boost investor interest and justify high valuations for companies such as Anthropic and OpenAI, which are both said to be preparing for large public listings. The article said some attendees argued that the rhetoric could be more about hype than safety.

Several prominent figures pushed back. Grindr chief executive George Arison told the BBC the comments were dangerous and said they led him to pause use of Anthropic tools at his company. Nvidia chief executive Jensen Huang also dismissed the extinction-risk claims, according to people present at the conference. Investors and analysts at the event also debated whether the warnings could prompt tighter regulation.

The broader context is that AI safety concerns have intensified as major labs release more powerful models and governments consider how to regulate them. Anthropic chief executive Dario Amodei separately called for slower development and stronger global rules, while the Trump administration has largely favoured rapid AI expansion for strategic reasons.